Modes of particle combustion in iron dust flames

The so-called argon/helium test is proposed to identify the combustion mode of particles in iron dust flames. Iron powders of different particle sizes varying from 3 to 34 μm were dispersed in simulated air compositions where nitrogen was replaced by argon and helium. Due to the independence of the particle burning rate on the oxygen diffusivity in the kinetic mode, the ratio between the flame speeds in helium and argon mixtures is expected to be smaller if the particle burning rate is controlled by reaction kinetics rather than oxygen diffusion. Experiments were performed in a reduced-gravity environment on a parabolic flight aircraft to prevent particle settling and buoyancy-driven disruption of the flame. Uniform suspensions of the iron powders were produced inside glass tubes and a flame was initiated at the open end of the tube. Quenching plate assemblies of various channel widths were installed inside the tube and pass or quench events were used to measure the quenching distance. Flame propagation was recorded by a high-speed digital camera and spectral measurements were used to determine the temperature of the condensed emitters in the flame. The measured flame speeds and quenching distances were in good agreement with previously developed one-dimensional, dust flame model where the particles are assumed to burn in a diffusive mode and heat losses are described on a volumetric basis. However, a significant drop of the ratio of flame speeds in helium and argon mixtures was observed for finer 3 μm particles and was attributed to a transition from the combustion controlled by diffusion for larger particles to kinetically controlled burning of micron-size particles. In helium mixtures, the lower flame temperatures measured in suspensions of fine particles in comparison to larger particles reinforces this assumption.     

Liquid flame: combustion of metal suspensions in liquid sulfur

Thermodynamic calculations show that some metals can react with sulfur without the formation of gaseous products at normal pressure and yet demonstrate sufficiently high flame temperatures to support the propagation of stable flames. For example, a stoichiometric ternary mixture of iron, manganese, and sulfur demonstrates gasless combustion at an equimolar concentration of iron and of manganese with an adiabatic flame temperature of about 2000 °C. Differential thermal analysis of the mixture shows no exothermic reactions below 280 °C. Therefore, sulfur in the mixture can be safely melted (m.p. 119 °C), converting a powder blend into a liquid suspension that is free from gas bubbles. Symmetrical cylindrical flames in shallow pools of suspensions of Fe and Mn powders in liquid sulfur and combustion of the same liquid mixtures in preheated narrow steel tubes have been studied to determine flame propagation speeds as a function of mixture composition. It was found that, contrary to the behavior of the calculated flame temperature, flame speed decreases with the increase of the manganese content in the mixture and is not affected by mixture dilution with the combustion product. Direct measurements of the flame temperatures by thermocouples indicated a weak dependence of the peak flame temperature on mixture composition and revealed a two-stage flame structure. The existence of the two distinct reaction zones in the mixture of two reactive metals with sulfur is in accordance with qualitative theoretical predictions by the theory of flame with parallel reactions existing in the literature. According to theory, the reaction with the higher flame speed in a corresponding binary single-metal–sulfur mixture will form the leading stage of the complex flame front and will govern the flame propagation speed in the ternary composition. The speed of flame propagation in pure Fe–S mixture is almost three times higher than the flame speed in Mn–S mixture. As a result, the iron–sulfur reaction dominates the flame propagation mechanism in Fe–Mn–S suspension.     

Flame behavior in heated porous sand bed

This paper reports experimental and numerical investigations on the combustion characteristics of a lean methane–air mixture in a heated porous sand bed. The porous bed consisted of sand (SiO₂) particles with a mean particle diameter of 0.56 mm. The horizontally placed quartz tube was heated externally to initiate the combustion reaction in the porous bed combustor. The stabilized flame location curve as a function of averaged mixture velocity was obtained for various equivalence ratios. Contrary to the earlier finding of a C-shape flame stabilization behavior, a new S-shape behavior was observed in the present study. This can be divided into three regimes: high, moderate, and low velocity regimes. In the low velocity regime, flame with very weak luminosity was confirmed and the stabilized flame location moved downstream with the increase of the mixture velocity. For the moderate velocity regime, a stable flat flame was observed and the flame location moved upstream with the increase of the mixture velocity. An oscillatory flame behavior was observed in the high velocity regime. In this oscillatory mode, the flame front oscillated with a characteristic time period of the order on 1 h and increased with the increase of the mixture velocity. In order to further understand these experimental results, one-dimensional computational studies with detailed chemistry and heat transfer mechanisms were carried out. The computational results were in good agreement with experimental observations. The computations showed that solid-to-solid radiation played a significant role in the flame stabilized location. From the examination of the flame structure, it was found that the flame behavior in the low velocity regime was similar to that of the flameless combustion mode.     

Combustion of bimodal nano/micron-sized aluminum particle dust in air

The combustion of bimodal nano/micron-sized aluminum particles with air is studied both analytically and experimentally in a well-characterized laminar particle-laden flow. Experimentally, an apparatus capable of producing Bunsen-type premixed flames was constructed to investigate the flame characteristics of bimodal-particle/air mixtures. The flame speed is positively affected by increasing the mass fraction of nano particles in the fuel formulation despite the lower flame luminosity and thicker flame zone. Theoretically, the flames are assumed to consist of several different regimes for fuel-lean mixture, including the preheat, flame, and post flame zones. The flame speed and temperature distribution are derived by solving the energy equation in each regime and matching the temperature and heat flux at the interfacial boundaries. The analysis allows for the investigation of the effects of particle composition and equivalence ratio on the burning characteristics of aluminum-particle/air mixtures. Reasonable agreement between theoretical results and experimental data was obtained in terms of flame speed. The flame structure of a bimodal particle dust cloud may display either an overlapping or a separated configuration, depending on the combustion properties of aluminum particles at different scales. At low percentages of nano particles in the fuel formulation, the flame exhibits a separated spatial structure with a wider flame regime. At higher nano-particle loadings, overlapping flame configurations are observed.     

Convective burning of fine ammonium nitrate–aluminum mixtures in a closed volume bomb

It is commonly assumed that the burning of ammonium nitrate–aluminum mixtures is much less prone to undergo a transition to explosion and detonation than similar mixtures based on ammonium perchlorate. However, this conclusion has been made for mixtures based on commercial-grade ammonium nitrate with large particles. In this study, the combustion of fine loose-packed mixtures of ammonium nitrate and aluminum in a closed-volume bomb has been examined. It has been shown that fine mixtures (ammonium nitrate with a particle size of less than 40 µm and an ASD-4 aluminum powder with spherical particles with a size of about 4 µm) undergo high-intensity combustion; in experiments with a stoichiometric mixture, explosions are observed. The explosions occur in the initial phase of convective combustion and lead to abrupt pressure pulsations with an amplitude of a few kilobars and to the destruction of the cup in which the sample is placed. The dynamics of development of the explosion has been analyzed in detail using numerical simulation. According to the results of experiments with varied parameters—the degree of dispersion of the ammonium nitrate powders, the aluminum content in the mixture, the length and diameter of the charge, and the level of pressure generated by the combustion of the igniter,—threshold conditions have been determined to separate the following modes: the absence of ignition, layer-by-layer combustion, or convective combustion with a transition into an explosion in experiments with a stoichiometric mixture.     

Stabilized flames in hybrid aluminum-methane-air mixtures

A premixed methane–air bunsen-type flame is seeded with micron-sized (d₃₂ = 5.6 μm) atomized aluminum powder over a wide range of solid fuel concentrations. The burning velocities of the resulting two-phase hybrid flame are determined using the total surface area of the inner flame cone and the known volumetric flow rate, and spatially resolved flame spectra are obtained with a spectral scanning system. Flame temperatures are derived through polychromatic fitting of Planck’s law to the continuous part of the spectrum. It is found that an increase in the solid fuel concentration changes the aluminum combustion regime from low temperature oxidation to full-fledged flame front propagation. For stoichiometric methane–air mixtures, the transition occurs in the aluminum concentration range of 140–220 g/m³ and is manifested by the appearance of AlO sub-oxide bands and an increase in the flame temperature to 2500 K. The flame burning velocity is found to decrease only slightly with an increase in aluminum concentration, in contrast to the rapid decrease in flame speed, followed by quenching, that is observed for flames seeded with inert SiC particles. The observed behavior of the burning velocity and flame temperature leads to the conclusion that intense aluminum combustion in a hybrid flame only occurs when the flame front propagating through the aluminum suspension is coupled to the methane–air flame.     

Turbulent flame speed and morphology of pure ammonia flames and blends with methane or hydrogen

Ammonia appears a promising hydrogen-energy carrier as well as a carbon-free fuel. However, there remain limited studies for ammonia combustion especially under turbulent conditions. To that end, using the spherically expanding flame configuration, the turbulent flame speeds of stoichiometric ammonia/air, ammonia/methane and ammonia/hydrogen were examined. The composition of blends studied are currently being investigated for gas turbine application and are evaluated at various turbulent intensities, covering different kinds of turbulent combustion regimes. Mie-scattering tomography was employed facilitating flame structure analysis. Results show that the flame propagation speed of ammonia/air increases exponentially with increasing hydrogen amount. It is less pronounced with increasing methane addition, analogous to the behavior displayed in the laminar regime. The turbulent to laminar flame speed ratio increases with turbulence intensity. However, smallest gains were observed at highest hydrogen content, presumably due to differences in the combustion regime, with the mixture located within the corrugated flamelet zone, with all other mixtures positioned within the thin reaction zone. A good correlation of the turbulent velocity based on the Karlovitz and Damköhler numbers is observable with the present dataset, as well as previous experimental measurements available in literature, suggesting that ammonia-based fuels may potentially be described following the usual turbulent combustion models. Flame morphology and stretch sensitivity analysis were conducted, revealing that flame curvature remains relatively similar for pure ammonia and ammonia-based mixtures. The wrinkling ratio is found to increase with both increasing ammonia fraction and turbulent intensity, in good agreement with measured increases in turbulent flame speed. On the other hand, in most cases, the flame stretch effect does not change significantly with increasing turbulence, whilst following a similar trend to that of the laminar Markstein length.     

Structure and stability characteristics of turbulent planar flames with inhomogeneous jet in a concentric flow slot burner

Turbulent flames with compositionally inhomogeneous mixtures are common in many combustion systems. Turbulent jet flames with a circular nozzle burner were used earlier to study the impact of inhomogeneous mixtures, and these studies showed that the nozzle radius affects the flame stability. Accordingly, planar turbulent flames with inhomogeneous turbulent jet are created in a concentric flow slot burner (CFSB) to avoid this effect in the present study. The stability characteristics, the mixing field structure, and the flame front structure were measured, and the correlations between stability and the mixing field structure were investigated. The mixture fraction field was measured in non-reacting jets at the nozzle exit using highly resolved Rayleigh scattering technique, and the flame front was measured in some selected turbulent flames using high-speed Planar Laser-Induced Fluorescence (PLIF) of OH technique. The data show strong correlations between flame stability and the range of mixture fraction fluctuations. The flames are highly stabilized within a mixing field environment with the range of fluctuation in mixture fraction close to the range of the flammability limits. The mixing field structure is also illustrated and discussed using a mixing regime diagram and showed that the scatter of the data of the different cases is consistent with the classified mixing regimes. Lean flames are stabilized in the current slot burner. The flame front structure topology varies consistently from thin, small curvature at the low level of turbulence and higher equivalence ratio to more wrinkled, larger curvature, but a thicker structure at a higher level of turbulence and lower equivalence ratio.     

Flame propagation speed and Markstein length of spherically expanding flames: Assessment of extrapolation and measurement techniques

Laminar burning velocities are of great importance in many combustion models as well as for validation and improvement of chemical kinetic schemes. Determining laminar burning velocities with high accuracy is quite challenging and different approaches exist. Hence, a comparison of existing methods measuring and evaluating laminar burning velocities is of interest. Here, two optical diagnostics, high speed tomography and Schlieren cinematography, are simultaneously set up to investigate methods for evaluating laminar flame speed in a spherical flame configuration. The hypothesis to obtain the same flame propagation radii over time with the two different techniques is addressed. Another important aspect is the estimation of flame properties, such as the unstretched flame propagation speed and Markstein length in the burnt gas phase and if these are estimated satisfactorily by common experimental approaches. Thorough evaluation of the data with several extrapolation techniques is undertaken. A systematic extrapolation approach is presented to give more confidence into results generated experimentally. The significance of the linear extrapolation routine is highlighted in this context. Measurements of spherically expanding flames are carried out in two high-pressure, high-temperature, constant-volume vessels at RWTH in Aachen, Germany and at ICARE in Orleans, France. For the discussion of the systematic extrapolation approach, flame speed measurements of methane / air mixtures with mixture Lewis numbers moderately away from unity are used. Conditions were varied from lean to rich mixtures, at temperatures of 298–373 K, and pressures of 1 atm and 5 bar.     

The effects of oxygen concentration and gas temperature on coal stream ignition and particle surface temperature in reducing-to-oxidizing environments

In the near-burner region of pulverized coal burners, two zones exist, with very different oxygen concentrations. The first zone is a locally reducing environment, caused by the fast release of volatiles from a region of dense coal particles, and the second zone, which is surrounding the first zone, is a hot oxidizing environment. The transition of coal particles from the reducing zone to the oxidizing zone affects early stage coal combustion characteristics, such as devolatilization, ignition and particle temperature history. In this work, we used a two-stage Hencken flat-flame burner to simulate the conditions that coal particles experience in practical combustors when they transition from a reducing environment to an oxidizing environments. The composition of the reducing environment was chosen to approximate that of a typical coal volatile. Three oxygen concentrations (5, 10 and 15 vol%) in the “ambient” oxidizing environment were tested, corresponding to those at different distances downstream from a commercial burner. The corresponding gas temperatures for the oxidizing environments were adjusted for the different oxygen concentrations such that the “volatile” flame temperatures were the same, as this is what would be expected in a commercial combustor. High speed videography was used to obtain the ignition characteristics, and RGB color pyrometry was used to measure particle surface temperatures. Two different sizes of coal particles were used. It is found that when particles undergo a reducing-to-oxidizing transition at high temperatures, the particles are preheated such that the critical factor for ignition delay is point at which the particle is in the presence of oxygen, not the concentration of oxygen. The ignition delay of large particles is found to be 53% longer than that of small particles due to their higher thermal mass and slower devolatilization. The oxygen concentration in the ambient have a negligible effect on early-stage particle temperatures.     

Measurements of the critical initiation radius and unsteady propagation of n-decane/air premixed flames

Unsteady flame propagation, the critical radius for flame initiation, and multiple flame regimes of n-decane/air mixtures are studied experimentally and computationally using outwardly propagating spherical flames at various equivalence ratios and pressures. The transient flame speeds, trajectories, and critical radius are measured. The experimental results are compared with direct numerical simulations using detailed high temperature kinetic models. Both experimental and numerical results show that there exist multiple flame regimes in the unsteady spherical flame initiation process. The transition between the flame regimes depends strongly on the mixture equivalence ratio (or Lewis number). It is found that there is a critical flame radius and that it increases dramatically as the mixture equivalence ratio and pressure decrease. The large increase of critical flame radius leads to a dramatic increase of the minimum ignition energy. Furthermore, the flame thickness and the radical pool concentration change significantly during the transition from the ignition flame regime to the self-sustained propagating flame regime. For the same steady state flame speeds, the predicted unsteady flame speeds and the critical flame radius differ significantly from the experimental results. Moreover, different chemical kinetic mechanisms predict different unsteady flame speeds. The existence of multiple flame regimes and the large critical radius of lean liquid fuel mixtures make the ignition of lean mixtures at low pressure and the development of a validated kinetic model more challenging. The unsteady flame regimes, speeds, and critical flame radius should be included as targets of future kinetic model development for turbulent combustion modeling.     

Experimental and kinetic study of 2,4,4-trimethyl-1-pentene and iso-octane in laminar flames

Laminar flame speeds of 2,4,4-trimethyl-1-pentene are investigated at equivalence ratios of 0.7–1.6, initial temperatures of 298–453 K and initial pressures of 0.1–0.5 MPa. The comparison between 2,4,4-trimethyl-1-pentene and iso-octane is also performed. Results show that 2,4,4-trimethyl-1-pentene has faster laminar flame speed than iso-octane. Chemical kinetic models (Metcalfe model, Modified model I) were tested against the present experimental data. The laminar flame speeds are apparently over-estimated by the Metcalfe model and under-predicted by the Modified model I. Therefore, high-level quantum mechanical calculations were used to revise the Modified model I to obtain Modified model II and it can give fairly good prediction at various conditions on laminar flame speeds. In addition, the chemical kinetic analysis was conducted. The analysis indicates both thermal and kinetic effects result in the discrepancy of laminar flame speeds between 2,4,4-trimethyl-1-pentene and iso-octane. Furthermore, IC₄H₈ plays a dominant role in laminar flame speeds of 2,4,4-trimethyl-1-pentene and iso-octane.     

直喷柴油机火焰浮起长度的数值模拟研究

采用三维CFD模型模拟了直喷柴油机缸内喷雾燃烧过程,模拟缸压曲线得到了实验的验证.通过高温区与喷嘴之间的稳定距离来确定柴油机火焰浮起长度,研究在不同进气条件下火焰浮起长度的变化情况.该模型成功地预测了火焰浮起长度随着初始进气压力的增大而减少,随着进气温度的升高出现先增大后减少的趋势.同时模拟了在不同EGR率下柴油机缸内燃烧情况,发现火焰浮起长度和燃料着火延迟时间都随EGR率的增加而增大.     

湿空气扩散燃烧火焰结构特性研究

利用二维粒子成像速度仪(PIV)对钝体燃烧器中的甲烷／湿空气扩散燃烧的速度场进行测量,考察其火焰的结构特性及其内部流动状况。通过对湿空气燃烧流场与普通燃烧流场的对比分析表明,湿空气燃烧情况下,两种燃烧状态的火焰(回流燃烧火焰和中心射流主导火焰)相互转换的燃空速度比(γ)值要比普通燃烧的小;湿空气燃烧使得喷嘴后的同流空气的速度降低,空气的回流作用减弱,燃料更容易冲出回流区,火焰的稳定性能变差。     

The effect of fuel inlet turbulence intensity on H2/CH4 flame structure of MILD combustion using the LES method

Large eddy simulations (LES) are employed to investigate the effect of the inlet turbulence intensity on the H₂/CH₄ flame structure in a hot and diluted co-flow stream which emulates the (Moderate or Intense Low-oxygen Dilution) MILD combustion regime. In this regard, three fuel inlet turbulence intensity profiles with the values of 4%, 7% and 10% are superimposed on the annular mixing layer. The effects of these changes on the flame structure under the MILD condition are studied for two oxygen concentrations of 3% and 9% (by mass) in the oxidiser stream and three hot co-flow temperatures 1300, 1500 and 1750 K. The turbulence-chemistry interaction of the numerically unresolved scales is modelled using the (Partially Stirred Reactor) PaSR method, where the full mechanism of GRI-2.11 represents the chemical reactions. The influences of the turbulence intensity on the flame structure under the MILD condition are studied by using the profile of temperature, CO and OH mass fractions in both physical and mixture fraction spaces at two downstream locations. Also, the effects of this parameter are investigated by contours of OH, HCO and CH₂O radicals in an area near the nozzle exit zone. Results show that increasing the fuel inlet turbulence intensity has a profound effect on the flame structure particularly at low oxygen mass fraction. This increment weakens the combustion zone and results in a decrease in the peak values of the flame temperature and OH and CO mass fractions. Furthermore, increasing the inlet turbulence intensity decreases the flame thickness, and increases the MILD flame instability and diffusion of un-burnt fuel through the flame front. These effects are reduced by increasing the hot co-flow temperature which reinforces the reaction zone.     

Effects of compression and stretch on the determination of laminar flame speeds using propagating spherical flames

The effects of flow compression and flame stretch on the accurate determination of laminar flame speeds at normal and elevated pressures using propagating spherical flames at constant pressure or constant volume are studied theoretically and numerically. The results show that both the compression-induced flow motion and flame stretch have significant impacts on the accuracy of flame speed determination. For the constant pressure method, a new method to obtain a compression-corrected flame speed (CCFS) for nearly constant pressure spherical bomb experiments is presented. Likewise, for the constant volume method, a technique to obtain a stretch-corrected flame speed (SCFS) at elevated pressures and temperatures is developed. The validity of theoretical results for both constant pressure and constant volume methods is demonstrated by numerical simulations using detailed chemistry for hydrogen/air, methane/air, and propane/air mixtures. It is shown that the present CCFS and SCFS methods not only improve the accuracy of the flame speed measurements significantly but also extend the parameter range of experimental conditions. The results can be used directly in experimental measurements of laminar flame speeds.     

Particle temperature and potassium release during combustion of single pulverized biomass char particles

This work investigated the combustion characteristics of single pulverized biomass-derived char particles. The char particles, in the size range 224–250 µm, were prepared in a drop tube furnace at pyrolysis temperatures of 1273 or 1473 K from four types of biomass particles – wheat straw, grape pomace, kiwi branches and rice husk. Subsequently, the char particles were injected upward into a confined region of hot combustion products produced by flat flames stabilized on a McKenna burner, with mean temperatures of 1460, 1580 and 1670 K and mean O₂ concentrations of 4.5, 6.5 and 8.5 vol%. The data reported include particle temperature, obtained using a two-color pyrometry technique, and potassium release rate, measured using a laser-induced photofragmentation fluorescence imaging technique. In addition, particle ignition delay time and burning time, obtained from the temporal evolution of the thermal radiation intensity of the burning char particles, are also reported. The results indicated that ignition of the char particles occurs simultaneously with the starting of the potassium release, then the particle burning intensity increases rapidly until it reaches a maximum, after which both the particle temperature and the potassium release rate remain approximately constant until the end of the char oxidation process. The char ignition process is temperature controlled, and the char oxidation process is oxygen diffusion controlled, with the total potassium release being independent of the oxygen concentration and the temperature of the combustion products. The combustion behavior of the chars studied is more affected by the char type than by the conditions used to prepare them.     

Model of heating and ignition of conductive polydisperse powder in electrostatic discharge

Heating of a conductive polydisperse powder by electrostatic discharge (ESD) is modelled numerically. Powder packing is described using a discrete element model; powder resistance is defined by geometry of particle contacts and properties of plasma produced by electrical breakdown between neighbour particles. A set of parametric calculations in combination with experimental data is used to determine necessary adjustable model parameters. The model predicts the temperature for each powder particle resulting from its heating by the ESD current. Location and packing of individual particles within the powder affects greatly their achieved temperatures and thus the likelihood of ignition. Consistently with experiments, a trend showing that smaller particles are generally heated to higher temperatures at a given ESD energy is detected for coarser powders; this trend becomes less clear for finer powders with particle sizes less than the breakdown distance given by the Paschen curve in air. Comparison of the experimental data and calculations suggests that the transition from single particle to cloud combustion occurs when the distance between the particles ignited by ESD becomes close to the flame size for the individual burning particle. This distance, inversely proportional to the number of ignited particles, is primarily determined by the ESD energy.     

Combustion waves in composite solid material of shell–core type

In this article, an asymptotic and numerical analysis of combustion wave propagation in shell–core composite solid energetic material is undertaken based on the diffusional–thermal model with an overall Arrhenius reaction step. Flame speed and structure are found for a broad range of parameter values. Two different regimes of flame propagation are identified. In the weak recuperation regime, the temperatures of the shell and core are monotonic functions of the coordinates, and they differ only slightly in the reaction zone of the flame. In the strong recuperation regime, the temperature of the shell is significantly higher than that of the core and has a sharp peak in the reaction zone with the maximum value exceeding the adiabatic flame temperature for pure energetic material. It is found that the highest level of flame acceleration in the composite material can be attained in the strong recuperation regime. The competition of these flame propagation regimes may lead to the coexistence of multiple combustion waves travelling with different velocities. The stability is investigated of combustion waves in the practically important strong recuperation regime.